We’ve reached a point where our DNA is no longer a static blueprint we simply inherit; it’s becoming a dynamic tool we use to navigate our lives. Entering 2026, the long-standing promise of genetics has moved out of the lab and into our daily routines. Groundbreaking new technologies are accelerating the pace with which we can deepen our understanding of DNA and improve our health. I’m incredibly excited about the convergence of several key trends that will not only enhance our understanding of ourselves, but also transform how we approach healthcare.
Embracing Artificial Intelligence
In genetics, as in almost every other aspect of our lives, one of the most transformative forces at play is Artificial Intelligence (AI). Imagine algorithms that can sift through vast datasets of genetic information, clinical records, and lifestyle factors with unparalleled speed and accuracy. AI is rapidly becoming an indispensable partner in identifying subtle genetic variations linked to disease, predicting individual responses to medications, and even designing novel therapeutic strategies.
At the 23andMe Research Institute, we’re combining our decades of genetics and machine learning/AI expertise to advance new techniques for genetics-based disease prediction. We recently presented some of this work at two conferences (NeurIPS 2025 and the Pacific Symposium on Biocomputing 2026) demonstrating new techniques for more accurate prediction of multiple diseases and for predicting an individual’s next most likely health event.
A More Complete Picture
The future of genetics is also multi-faceted, quite literally. We’re increasingly embracing “multi-omics” studies, which combine genetic data (genomics) with information from other biological molecules like proteins (proteomics), metabolites (metabolomics), and other kinds of data. Genetics and proteomics, for example, have been shown to provide different kinds of information. Bringing them together allows us to better understand what might cause a disease or develop better drugs compared to just looking at genetics alone. Think of it like assembling a complete puzzle rather than just looking at one piece. By integrating these diverse layers of biological information, we gain a much richer and more dynamic picture of an individual’s health, disease risk, and response to treatment. This holistic view is crucial for understanding complex conditions and developing more effective, targeted interventions.
Bridging the Gap to the Clinic
In the clinic, we anticipate a significant expansion in the applications of Polygenic Risk Scores (PRS). While individual genetic variants can have a small impact, PRS combine the effects of many common genetic variants across the genome to estimate an individual’s risk for common diseases like heart disease, type 2 diabetes, and certain cancers. PRS can be used in many ways, including in clinical trials to help recruit and identify the individuals who may benefit the most from a potential medication. These scores have the power to help individuals and their healthcare providers with actionable insights, enabling earlier preventative measures and more tailored screening programs.
Clinical applications of PRS have often been hampered by lack of studies on the clinical validity or utility of such scores. However, more and more researchers are investigating how people actually use their genetic information with their healthcare professionals including some of us at the 23andMe Research Institute. Recently we published a study showing that genetic information on depression can be delivered responsibly, empowering individuals with valuable knowledge without unnecessarily fueling anxiety or depression. In 2026, we plan to continue research into the clinical utilization of personal genetic information.
Secure data sharing
Finally, at the heart of all these advancements lies the critical need for secure and responsible data sharing. Genetic research often requires large numbers of research participants to contribute their data, and new discoveries rarely happen in isolation. Collaboration between scientists and sharing of data (always with participant consent) is critical for our collective understanding of the human genome.
At the 23andMe Research Institute, we are deeply committed to maintaining the highest standards of privacy and security, ensuring that participant data is handled with the utmost care and used ethically to drive scientific progress. Our research is reviewed by Salus IRB, an independent external AAHRPP-accredited Institutional Review Board (IRB), and members can change whether they consent to participate in research or not at any time. By fostering secure data sharing frameworks, we can collectively unlock the full potential of this information, amplify the reach and impact of our participants’ contributions, and ultimately, benefit humanity as a whole.
The journey ahead is exciting, and with the continued collaboration of researchers, participants, and cutting-edge technology, we are confident that the next few years will usher in an unprecedented era of genetic understanding and improved health for everyone.
Learn more about 23andMe Research
Read on to learn about some of the new discoveries and more than a dozen new research papers published in 2025, adding to the more than 300 total publications by 23andMe Research Institute and its collaborators. This is only possible because of the continued participation of millions of consented 23andMe research participants.
2025 Research Recap: Advancing Genetic Innovation
Mental Health and Neurodevelopment
Obsessive-Compulsive Disorder
The largest ever genetic study for Obsessive-Compulsive Disorder (OCD) identified 30 independent genetic variants associated with the condition, with half of them being entirely new discoveries. Consented 23andMe research participants made up about half of the total cohort, demonstrating the power of large-scale data. Researchers found the genetic signals for OCD were highly enriched in six types of human brain tissue, with no enrichment outside the brain. This work also showed correlations between OCD and other conditions like anxiety, depression, and Tourette syndrome. Understanding this genetic overlap may be key to understanding mental health as a whole.
Strom, Nora I et al. “Genome-wide analyses identify 30 loci associated with obsessive-compulsive disorder.” Nature genetics vol. 57,6 (2025): 1389-1401. doi:10.1038/s41588-025-02189-z
Depression Genetics and Treatment
An international, groundbreaking study analyzed over five million genomes (more than a third of which were from consenting 23andMe research participants) to uncover nearly 700 genetic variations linked to depression, identifying 308 specific genes. Importantly, the study included diverse participants from African, East Asian, Hispanic, and South Asian backgrounds, leading to the discovery of 100 new genetic variants previously missed in European-only studies. These findings offer insights into the biological underpinnings of depression and potential new treatment options.
Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium. “Trans-ancestry genome-wide study of depression identifies 697 associations implicating cell types and pharmacotherapies.” Cell vol. 188,3 (2025): 640-652.e9. doi:10.1016/j.cell.2024.12.002
Separately, researchers at Johnson & Johnson and the 23andMe Research Institute successfully developed a predictive model using self-reported data from 23andMe participants to accurately identify individuals at high risk for treatment-resistant depression. The advanced machine learning model utilized 101 predictive factors (excluding genetics) to predict those at high risk 78% of the time. This potentially provides a tool to flag those who might not respond to standard therapies, which may accelerate the process of finding alternatives and is a massive breakthrough for those suffering with treatment-resistant depression.
Vairavan, Srinivasan et al. “Prediction of treatment-resistant depression using the 23andMe survey data.” Scientific reports vol. 15,1 33881. 30 Sep. 2025, doi:10.1038/s41598-025-08460-x
Finally, addressing lingering apprehension about personalized genetic testing, a 23andMe study found that receiving a high genetic predisposition report for depression did not increase feelings of depression or anxiety, suggesting this information can be delivered responsibly, and can empower individuals with valuable knowledge without unnecessarily fueling anxiety or depression.
Berns, Rebecca M K et al. “Viewing Direct-to-Consumer Genetic Test Results for Depression Risk Is Psychologically Well Tolerated: Evidence from a Longitudinal Equivalence Study.” HGG advances, 100553. 8 Dec. 2025, doi:10.1016/j.xhgg.2025.100553
Dyslexia and Reading Ability
The largest genetic study on dyslexia to date combined data from a 23andMe genome-wide association study (GWAS) on dyslexia with data from the GenLang Consortium to identify 80 independent places in the genome that are significantly associated with dyslexia, with 13 of these associations being completely novel discoveries. These findings point to genes active in neuronal pathways during early brain development. Identifying the biological pathways involved in dyslexia helps open future research into why reading is difficult for some people and how to develop better support strategies.
Mountford, Hayley S et al. “Multivariate genome-wide association analysis of dyslexia and quantitative reading skill improves gene discovery.” Translational psychiatry vol. 15,1 289. 18 Aug. 2025, doi:10.1038/s41398-025-03514-0
A separate study involving a dyslexia polygenic score, developed in part thanks to 23andMe research participants, determined that genetic and socioeconomic factors both contribute independently to reading and spelling performance in Australian and UK populations in contrast to what was found in earlier US-based studies. This difference in findings depending on the country studied is an important reminder for scientists to consider how different environments may change how genetics can influence complex traits such as reading ability.
Bicona, Diana et al. “Dyslexia Polygenic Index and Socio-Economic Status Interaction Effects on Reading Skills in Australia and the United Kingdom.” Behavior genetics vol. 55,5 (2025): 395-406. doi:10.1007/s10519-025-10230-4
Stuttering
A new study powered by 23andMe data identified 57 genetic variants linked to self-reported stuttering. The results found genetic overlap with traits such as depression, autism, and beat synchronization. This research shows that genetic risk for stuttering is complex involving many genes, and it opens up many new avenues of research into this common, complex condition.
Polikowsky, Hannah G et al. “Large-scale genome-wide analyses of stuttering.” Nature genetics vol. 57,8 (2025): 1835-1847. doi:10.1038/s41588-025-02267-2
Delay Discounting
Research into the genetics of delay discounting (the tendency to devalue larger future rewards in favor of smaller more immediate rewards) utilizing data from over 134,000 consented 23andMe research participants, uncovered 14 distinct genetic variants. The genetics of delay discounting are also associated with many complex behaviors, such as smoking, as well as several health conditions. This research sets up future studies aimed at discovering better prevention strategies, diagnostic tools, and treatments.
Thorpe, Hayley H A et al. “Genome-wide association study of delay discounting identifies 11 loci and reveals transdiagnostic associations across mental and physical health.” Molecular psychiatry, 10.1038/s41380-025-03356-8. 25 Nov. 2025, doi:10.1038/s41380-025-03356-8
Cancer Research
Genetic Risk and Cancer Survival
A collaboration between the 23andMe Research Institute and the Dana-Farber Cancer Institute made the surprising finding that breast cancer patients with a high polygenic risk score for breast cancer (meaning the impact of many genetic variants added up to a higher risk for breast cancer) may experience longer average survival times than those with low scores. Genetic variants come in two types: those you’re born with (germline mutations) that appear in every cell, and those that develop over time in specific tissues (somatic mutations). The research suggests tumors driven by genetic changes you’re born with (measured by PRS) may be less aggressive than those caused by later-in-life somatic mutations.
Kurant, Danielle E et al. “Association between polygenic risk and survival in breast cancer patients.” BMC cancer vol. 25,1 1393. 28 Aug. 2025, doi:10.1186/s12885-025-14640-9
Lung Cancer Genetics Study (Ongoing)
Launched in 2024, the study is an ongoing collaboration with the Susan Wojcicki Foundation and over 20 advocacy groups. This year the 23andMe Research Institute, in partnership with Troper Wojcicki Philanthropies and Lifebit, established an open-source data platform to provide qualified researchers with no-cost access to de-identified genetic and clinical data from consented participants.
Early data from this study was presented at the IASLC 2025 World Conference on Lung Cancer. Researchers found no correlation between the presence of rare genetic variants and smoking history among 1,010 individuals with lung cancer. This means that in this group of research participants, rare genetic variants impact lung cancer risk the same in people regardless of their smoking status. As the number of participants in the Lung Cancer Genetics Study grows, researchers aim to reassess and continue to look for genetic variants that may influence the development of lung cancer. Learn more about the myths about lung cancer, and how patient stories can help shed light on the need for more research.
Learn more about the Lung Cancer Genetics Study
Metabolic Health and Other Discoveries
BMI and Polygenic Scores
A collaboration combining 23andMe and GIANT consortium data created the most predictive polygenic score for BMI to date using up to 5.1 million individuals from diverse backgrounds. The study linked a higher genetic predisposition to a high BMI to faster childhood weight gain, suggesting early intervention opportunities.
Smit, Roelof A J et al. “Polygenic prediction of body mass index and obesity through the life course and across ancestries.” Nature medicine vol. 31,9 (2025): 3151-3168. doi:10.1038/s41591-025-03827-z
Genetics of Cannabis Use
Thanks to over 130,000 consented 23andMe research participants, researchers were able to reveal genetic links to both lifetime cannabis use (associated with genes CADM2 and GRM3) and the frequency of use. This work established genetic overlap between cannabis use and psychiatric, cognitive, and physical health outcomes, supporting the value of investigating pre-addiction traits, which may eventually help predict risk for cannabis use disorder and related health conditions.
Thorpe, Hayley H A et al. “Genome-wide association studies of lifetime and frequency of cannabis use in 131,895 individuals.” Molecular psychiatry, 10.1038/s41380-025-03219-2. 13 Oct. 2025, doi:10.1038/s41380-025-03219-2
